DOI 10.1393/ncc/i2013-11602-7
Colloquia: QCDN12
IL NUOVO CIMENTO Vol. 36 C, N. 5 Settembre-Ottobre 2013
The Drell-Yan measurement at COMPASS-II
S. Takekawaon behalf of the COMPASS Collaboration
INFN, Sezione di Torino - via P. Giuria 1, 10125 Torino, Italy ricevuto il 18 Aprile 2013
Summary. — The Drell-Yan process can be used to access Transverse Momentum Dependent Parton Distribution Functions, such as the Boer-Mulders and the Sivers functions, and the transversity function, providing complementary information to what is known from Semi Inclusive Deep Inelastic Scattering data. The COM-PASS experiment offers the possibility to study single polarized Drell-Yan processes (π−+ p↑ → μ+μ−+ X), making use of its large acceptance spectrometer and its unique transversely polarised proton target. A fundamental test of the factorization theorem in the non-perturbative QCD can be performed as well, by verifying the sign change of the Boer-Mulders and the Sivers functions depending if they are accessed via SIDIS or Drell-Yan process. The future Drell-Yan experiment at COMPASS is foreseen in late 2014.
PACS 13.85.Qk – Drell-Yan process. PACS 14.20.Dh – Nucleons.
PACS 12.38.Bx – Perturbation theory applied to quantum chromodynamics.
1. – Introduction
The COMPASS Collaboration has been contributing to the knowledge of the spin structure of the nucleon by studying Semi Inclusive Deep Inelastic Scattering (SIDIS) of muons on longitudinally and transversely polarised nucleons. The new COMPASS-II Proposal [1] extends the COMPASS physics program by exploring two new processes to investigate the nucleon spin structure: Deeply Virtual Compton Scattering (DVCS) and polarised Drell-Yan. After the approval of the CERN Research Board on 1st December 2010, a preliminary schedule foresees the start of the Drell-Yan program at COMPASS in late 2014.
2. – The Drell-Yan process and TMDs
When one permits non-zero intrinsic transverse momentum kT of partons inside the nucleons, at leading order, the proton can be described by eight Transverse Momen-tum Dependent Parton Distribution Functions (TMD PDFs). They are: f1(x, k2T),
c
248 S. TAKEKAWA on behalf of the COMPASS COLLABORATION
g1L(x, k2T), h1(x, kT2), g1T(x, k2T), h⊥1L(x, kT2), h⊥1T(x, k2T), h⊥1(x, k2T) and f1T⊥(x, k2T), where x is the fraction of the proton momentum carried by the partons. The TMD PDFs describe the correlation of momentum and spin of partons with the spin of the parent proton if it is either unpolarised or longitudinally/transversely polarised, so their study can help to understand the helicity structure of nucleon and to solve the long lasting spin puzzle. The first three functions listed above, when integrated over kT, yields to themomentum distribution, the helicity distribution and transversity function, commonly known in the collinear approximation. The TMD PDFs, like the Pretzelosity
h⊥1T(x, k2
T), the Boer-Mulders h⊥1(x, k2T) and the Sivers f1T⊥(x, k2T) functions, can be ac-cessed studying the Semi Inclusive Deep Inelastic (SIDIS) scattering and the Drell-Yan process. The Boer-Mulders and the Sivers functions are na¨ıvely T-odd; as a conse-quence, QCD factorization predicts that both functions should change their sign when measured in SIDIS experiments and in Drell-Yan experiments [2], thus an important test of QCD can be performed, since the relation should hold: f1T⊥(DY ) = −f1T⊥(SIDIS) and h⊥1(DY ) =−h⊥1(SIDIS).
The Drell-Yan process is the annihilation of a quark-antiquark pair coming from two hadrons into a virtual photon which produces a pair of leptons [3]. The angular distribution of the lepton was widely studied in the past. One of the results is the so-called Lam-Tung sum rule [4], which says 1− λ = 2ν. λ, μ and ν are three parameters which appears in the expression of the unpolarised angular distribution of the leptons:
(1) 1 σ dσ dΩ = 3 4π 1 λ + 3
1 + λ cos2θ + μ sin2θ cos φ +ν
2 sin
2θ cos 2φ.
In the collinear approximation, the QED predicts λ = 1 and μ = ν = 0. However mea-surements were not in agreement with the QED prediction [5, 6], with non-zero values for μ and ν, and even the validity of the sum rule was put in doubt. The distributions of azimuthal angles between the lepton and hadron planes, as well as between the lepton plane and the spin of the polarised hadron, provide information on the TMD PDFs in the Drell-Yan process [7]. In case of the reaction π−+ p↑ → μ+μ− + X (single po-larised Drell-Yan), with a pion beam scattering on a (popo-larised) proton, four asymmetry terms can be isolated to access TMDs of the two hadrons in the angular distribution is [1] dσ d4qdΩ = α2 em F q2σˆU 1 + D[sin2θ]Acos 2φ U cos 2φ (2) +|ST| Asin φS T sin φS+ D[sin2θ] Asin(2φ+φS) T sin(2φ + φS) +Asin(2φT −φS)sin(2φ− φS) ,
where D[f (θ)]are the depolarisation factors, ST is the target polarisation vector, F is the flux of incoming hadrons, the As are the azimuthal asymmetries, ˆσU is the part of the cross section which survives the integration over φ and φS, φ and θ are the azimuthal and polar angles of the lepton in the Collins-Soper frame [8] and φS is the angle of the spin in the proton rest frame. Each asymmetry term gives access to the convolution of two TMD PDFs: in Acos 2φU of the Boer-Mulders functions of the incoming hadrons, in
Asin φS
THE DRELL-YAN MEASUREMENT AT COMPASS-II 249
Fig. 1. – Comparisons of several prediction of the asymmetry term Asin φS
T in the COMPASS kinematic range and 4–9 GeV/c2 virtual photon mass range: black solid and dashed [11],
dot-dashed [12], red [13], boxes [14], green short dot-dashed [15]. In blue points with error bars coming from a two bin analysis and two years of data taking.
Asin(2φ+φS)
T of the Boer-Mulders function of the beam hadron and to the Pretzelosity function of the target nucleon, in Asin(2φT −φS) of the Boer-Mulders function of the beam hadron and the transversity function of the target nucleon.
3. – The COMPASS spectrometer
COMPASS is a fixed target experiment located in the North Area at CERN [9, 10]. The COMPASS spectrometer is built at the end of the M2 beam line which can deliver a hadron beam to the polarised target. The spectrometer has a large acceptance, up to ±180 mrad due to its two stages structure. Each stage is built around an analysing magnet and it is equiped with tracking detectors (GEM, MicroMegas, MWPC, Straws, SciFi and Silicon detectors) and hadronic and electromagnetic calorimeters. The first stage has also a RICH detector for particle identification while muon filters provide muon identification in both stages. The unique polarised target is capable to polarise ammonia (proton target) up to 90% with a dilution factor of 0.22 (for the Drell-Yan process). The COMPASS spectrometer gives the possibility to study the Drell-Yan process in the reaction π−+ p↑ → μ+μ− + X. Minor upgrades of the COMPASS spectrometer are required to make it efficient for the future Drell-Yan experiment. The most important one is the installation of a hadron absorber after the target, to significantly reduce the secondary particle flux and to suppress the background from the pion decay. In fact, a high intensity beam is needed to collect a high luminosity to compensate the small cross-section of the Drell-Yan process. The absorber will be made of aluminum oxide (Al2O3) with a tungsten core to stop the beam which does not interact in the target. The target platform will be moved upstream along the beam line to make enough room to place the hadron absorber. Monte Carlo simulations and particle flux studies were done to get a design which fulfils physics requirements, as well as radioprotection rules and engineering needs. Simulations allowed to study the acceptance of the COMPASS spectrometer to Drell-Yan events. The study was done looking at the mass region of the virtual photon between 4 and 9 GeV/c2, the so-called “safe region” which is almost background free compared to the region below 3.5 GeV/c2.
250 S. TAKEKAWA on behalf of the COMPASS COLLABORATION
4. – Beam tests
In the last few years, short Drell-Yan beam tests were performed at COMPASS. At the end of 2009, an important test was done in the conditions close to the future Drell-Yan measurement with a hadron absorber. Several results were achieved: the feasibility of the measurement was proven, the yeilds of J/ψ and Drell-Yan events were measured and the possibility to separate the Drell-Yan events coming from two different target cells was shown. During the test, radioprotection limits were respected, the detector occupancies were low and a good agreement was found between Monte Carlo simulations and reality.
5. – Predictions
Single spin asymmetries are expected to be quite large in the proton valence region, which is covered by the COMPASS acceptance to Drell-Yan events. Several predictions are available for the COMPASS kinematic and 4–9 GeV/c2 virtual photon mass range: fig. 1 shows the projections of the statistical error on the Sivers asymmetry as compared to the various theoretical predictions, corresponding to two years of data taking. These predictions do not take into account TMD evolution.
6. – Conclusion
The COMPASS experiment has the possibility to give new results in the under-standing of the physics of the spin of the proton using single polarised Drell-Yan process, probing the TMD PDFs. An important test of the TMD factorization ap-proach in QCD can be performed by checking whether the fondamental QCD prediction (f1T⊥(DY ) =−f1T⊥(SIDIS) and h⊥1(DY ) =−h⊥1(SIDIS)) is correct or not and if a
gen-eralized universality of TMD PDFs exists. The preliminary schedule of the COMPASS-II
program foresees the start of the Drell-Yan measurent in late 2014.
REFERENCES
[1] Gautheron F. et al., COMPASS-II Proposal, 2010, SPSC-P-340. [2] Collins J. C., Phys. Lett. B, 536 (2002) 43.
[3] Drell S. D. and Yan Tung-Mow, Phys. Rev. Lett., 25 (1970) 316. [4] Lam C. S. and Tung Wu-Ki, Phys. Rev. D, 18 (1978) 2447. [5] Falciano S. et al., Z. Phys. C, 31 (1986) 513.
[6] Conway J. S. et al., Phys. Rev. D, 39 (1989) 92.
[7] Arnold S., Metz A. and Schlegel M., Phys. Rev. D, 79 (2009) 034005. [8] Collins J. C. and Soper Davison E., Phys. Rev. D, 16 (1977) 2219.
[9] Baum G. et al., COMPASS: A Proposal for a Common Muon and Proton Apparatus for Structure and Spectroscopy, CERN-SPSLC-96-14.
[10] Abbon P. et al., Nucl. Instrum. Methods A, 577 (2007) 455.
[11] Efremov A. V., Goeke K., Menzel S., Metz A. and Schweitzer P., Phys. Lett. B, 612 (2005) 233.
[12] Collins J. C. et al., Phys. Rev. D, 73 (2006) 014021. [13] Anselmino M. et al., Phys. Rev. D, 79 (2009) 054010.
[14] Bianconi Andrea and Radici Marco, Phys. Rev. D, 73 (2006) 114002.
[15] Bacchetta Alessandro, Conti Francesco and Radici Marco, Phys. Rev. D, 78 (2008) 074010.
